On May 7, 2026, Google quietly added a deprecation banner to its FAQ structured data developer documentation. Within hours, half the SEO industry declared FAQ schema dead. The other half started arguing it was now more important than ever for AI search. Both reactions miss the point.
The rich result is gone. The schema type is not. Those two things are different, and the difference decides what you should actually do this week.
This guide cuts through the noise. You will learn what changed, what it means for your existing FAQ pages, where FAQ structured data still pulls weight in 2026, and the exact decisions to make before Google retires Search Console reporting in June and the API in August.
What FAQ Structured Data Actually Is
FAQ structured data is JSON-LD markup, defined under the schema.org FAQPage type, that tells search engines and AI systems: “this page contains a list of questions and answers, and here is the exact pairing.” Each question gets a Question object, each answer gets an acceptedAnswer object, and the whole thing sits in a script tag in your page source.
It is not the same as having an FAQ section. The FAQ section is the visible block on your page. The structured data is the machine-readable label wrapped around it.
That distinction has always mattered. After May 2026, it matters more.
There is also a separate but related type called QAPage. Use QAPage when one question has multiple user-submitted answers (forums, community Q&A). Use FAQPage when the site owner provides the single official answer. People conflate them constantly, and the wrong choice will get the markup ignored.
How Google Changed FAQ Rich Results
The deprecation didn’t happen overnight. It was a three-year wind-down.
- May 2019: Google launched FAQ rich results. Within months, sites that added the schema saw their listings expand vertically in the SERP, often capturing three to four times the screen space of unmarked competitors.
- August 2023: Google restricted FAQ rich results to “well-known, authoritative government and health websites” only. Across tracking tools, FAQ rich result impressions for everyone else collapsed within weeks.
- March 2026: Tracking data showed FAQ rich result impressions dropping again, even for the remaining eligible sites.
- May 7, 2026: Google ended the feature entirely. The banner on the documentation reads: “FAQ rich results are no longer appearing in Google Search. We will be dropping the FAQ search appearance, rich result report, and support in the Rich results test in June 2026. To allow time for adjusting your API calls, support for the FAQ rich result in the Search Console API will be removed in August 2026.”
Google didn’t publish a blog post explaining why. No reasoning was given beyond the deprecation notice added to the documentation. The pattern matches what Google did with HowTo rich results in late 2023: restrict eligibility, then retire the feature entirely.
For more context on how Google has been steadily trimming rich result features, see our breakdown of the broader rich results update and what SEOs should do now.
Google FAQ Rich Results Eligibility in 2026
Short version: nobody is eligible anymore. Not authoritative government sites, not health sites, not anyone.
Between August 2023 and May 2026, eligibility was technically open to a narrow band of government and health domains. As of the May announcement, the feature itself stopped firing across all categories. The eligibility question is now moot, not because the rules got stricter, but because the door is closed.
A few practical points that confuse people:
- The schema type is not deprecated. FAQPage remains a valid schema.org type. Your markup will not throw errors, will not produce manual actions, and will not damage your rankings if you leave it in place.
- Other search engines may still process it. Bing, DuckDuckGo, and various AI crawlers continue to parse FAQ markup. Google removing its rich result is a Google decision, not an industry-wide one.
- The Search Console FAQ enhancement report is on a timer. It disappears in June 2026, along with the Rich Results Test FAQ check. The Search Console API stops returning FAQ data in August 2026.
If your team built monitoring dashboards around FAQ enhancement data, you have a narrow window to export historical numbers before they vanish. Our guide to reading Google Search Console reports covers the export workflows you can use to capture that data while it is still available.
Google FAQ Rich Results Guidelines: What They Still Say
The Google FAQ structured data documentation hasn’t been wiped, it’s been topped with a deprecation banner, with the underlying guidelines kept intact. That matters because the guidelines still describe what valid FAQPage markup looks like, and other systems beyond Google still apply similar rules. The core requirements have not changed:
- Use FAQPage only when you provide the single authoritative answer to each question. Use QAPage for community-driven Q&A.
- The question text in the structured data must match the question shown on the page. Same for the answer.
- Don’t hide questions or answers from users while marking them up for crawlers. Cloaked schema is a manual action risk regardless of whether a rich result fires.
- No promotional content in the answer. No advertisements. No “buy now” calls to action wrapped in Answer markup.
- Each question gets one answer. If you have several official answers, restructure the page.
These rules survived the deprecation because they describe what FAQ content should actually look like. The visual feature went away. The content quality bar did not.
Does FAQ Schema Still Matter for SEO?
The honest answer is layered, so here it is in three parts.
For traditional Google rankings: No direct lift. Google has repeatedly said FAQ markup is not a ranking factor, and with no rich result to earn, you cannot use it to expand your SERP footprint either. Adding FAQ schema to a product page in 2026 will not move you from position 6 to position 3.
For Google AI Overviews and AI Mode: Unclear. Google has not connected FAQ structured data to AI Overview inclusion, and Glenn Gabe, president of G-Squared Interactive, observed after the deprecation that the company’s actions suggest FAQ markup is not weighted as an AI surface signal. Pages do get cited in AI Overviews without any FAQ schema at all. If you want to understand what actually drives AI citation eligibility, our breakdown of the Google Search AI Overview impact on SEO walks through the patterns we’ve seen across client sites.
For other AI search platforms: Possible signal, not a guaranteed one. ChatGPT search, Perplexity, and Bing Copilot ground their answers partly in web content their crawlers parse. Structured Q&A pairs are easier to extract than unstructured prose, which is the case some practitioners make for keeping FAQ schema on key pages. The honest caveat: no major AI platform has publicly documented how it weights FAQPage markup, and a clean question heading with a self-contained prose answer often performs the same job.
The trap to avoid is treating FAQ schema as a magic shortcut to AI visibility. The same templated FAQ patterns that got the rich result killed are unlikely to earn citations in any answer engine. The structured data layer matters less than the underlying content quality. For a deeper look at how AI engines actually pull from structured pages, our guide to the role of structured data in answer engine optimization covers what we’ve seen drive citation lift.
Best Practices for Implementing FAQ Structured Data Today
If you’re keeping FAQ schema on a page in 2026, the bar is higher than it used to be. Decorative markup that mirrors competitor templates does nothing for you. Here is what still works.
Match the schema to real content on the page. Every question and answer in the JSON-LD must appear in visible HTML. No hidden questions injected only for crawlers.
Pick questions your audience actually asks. Pull them from support tickets, sales call recordings, Reddit threads about your topic, and the People Also Ask box. Invented questions written to hit keywords don’t earn citations and don’t help users.
Keep answers complete but short. A two- to four-sentence answer that fully resolves the question is the right shape for both human readers and AI extraction. Long, hedged paragraphs dilute the signal.
Use one FAQPage block per URL. Google’s documentation is explicit: only one FAQPage definition per page. If you have multiple Q&A clusters, nest them all inside a single mainEntity array.
Avoid promotional language inside Answer fields. “Our award-winning team” or “the best service in the industry” is the kind of content that historically broke eligibility and is still flagged by structured data validators.
Validate before deploying. Use the schema.org validator or any third-party JSON-LD linter. The Google Rich Results Test will lose its FAQ check in June 2026, so workflows that depend on it need updating.
Keep schema and content versioned together. When you change a visible FAQ answer, change the JSON-LD in the same commit. Drift between the two is the most common cause of long-term schema problems.
A minimal valid FAQPage block looks like this:
That’s the entire pattern. Extra properties add complexity without much upside in 2026.
Common Mistakes to Avoid
Most of the FAQ schema problems we still see fall into the same handful of patterns. The deprecation makes some of these worse, not better.
- Treating FAQ schema as an AI search shortcut. Lily Ray, VP of SEO Strategy and Research at Amsive, documented one of the leading theories behind the May 2026 deprecation: a flood of generic “add FAQ schema for AI visibility” advice produced templated markup across what she counted as roughly 168,000 articles, much of it not describing genuine page content. Google’s historical pattern, going back to 2019, is to retire features that get aggressively scaled and stop faithfully describing pages.
- Adding the same FAQ block to every page. Duplicated FAQ markup across a site adds no value and signals templated content. If five service pages all share the same five questions, you have one FAQ section and four redundant blocks.
- Marking up content the user can’t see. Schema describing collapsed accordion content that is loaded client-side without a clear path for the crawler to reach it is a recurring issue. If a user has to click to see the answer, that’s fine. If the rendered HTML never contains the answer, that’s a problem.
- Generating FAQ schema from H2 headings automatically. Some plugins do this. The output looks valid but the H2s aren’t necessarily questions, and the paragraphs underneath aren’t necessarily answers. Auto-generation without editorial review produces noise.
- Forgetting to remove the old FAQ reporting integrations. If your dashboards pull FAQ data from the Search Console API, that pipeline breaks in August 2026. Audit your reporting stack now.
What the SEO Community Is Actually Saying
The conversation across LinkedIn, X, and industry Slack groups in the week after the deprecation broke into a few clear camps. None of them have it entirely right, but the patterns are worth understanding because they shape what your stakeholders will probably ask about next.
- Camp one: “Schema is dead.” This is the loudest take and the most wrong. The rich result is dead. Structured data as a category is still actively supported by Google for Product, Article, LocalBusiness, Event, Review, Organization, Person, and many others. Treating FAQ deprecation as evidence that all schema is on the way out misreads Google’s pattern.
- Camp two: “FAQ schema matters more than ever for AI.” This is the take that probably contributed to the deprecation. The argument is that AI systems extract Q&A pairs more reliably when they’re explicitly marked up. The argument is plausible but underdocumented. No major AI platform has published weighting details, and the same outcome can usually be achieved with clear prose headings.
- Camp three: “Remove all FAQ schema immediately.” Also wrong. Google has explicitly stated that unused structured data does not cause Search issues. There is no penalty for leaving valid FAQPage markup in place. Rushing to remove it across thousands of pages costs engineering time for zero measurable benefit.
- Camp four: the quiet majority. A practical read that says: keep it where the content is genuinely Q&A-shaped, remove it where it was added decoratively, and stop treating it as a primary SEO tactic. This is the take that ages well.
For broader context on the AI search shift driving so much of this discussion, our analysis of why Google rankings mean less in AI search covers the strategic frame most teams are missing.
Step-by-Step: What to Do With Your FAQ Schema This Quarter
Here’s the order of operations we’re recommending to clients. It maps to the Google deprecation timeline so you don’t miss the reporting export windows.
- Step 1: Export your historical FAQ rich result data. Before June 2026, log into Search Console and pull the FAQ enhancement report for the last 16 months. Save it as CSV. If you use the API for automated reporting, capture the historical data before August. You won’t get it back.
- Step 2: Inventory which pages have FAQ schema. Most teams don’t actually know. Run a site crawl with Screaming Frog, Sitebulb, or any crawler that parses JSON-LD. Output: a list of every URL with a FAQPage block.
- Step 3: Categorize each page. For each URL with FAQ markup, decide:
- Keep: The page is genuinely Q&A-shaped (support articles, help center entries, regulator pages, guides where users arrive with specific questions). The schema accurately describes content visitors actually use.
- Audit: The FAQ section exists but the questions are stale, low-quality, or feel templated. Mark for an editorial rewrite, not removal.
- Remove: The FAQ block was added purely for the SERP feature. The questions don’t reflect real user intent. Remove the markup and consider removing the section itself if it’s not serving readers.
- Step 4: Update your reporting workflows. Strip FAQ rich result dependencies from dashboards. Replace with metrics that still work, overall organic impressions, total clicks, page-level CTR. If you track AI citation visibility separately, that’s the new lane.
- Step 5: Refocus the schema effort. The hours you used to spend optimizing FAQ markup are better spent on schema types that still produce rich results: Product, Article, Organization, LocalBusiness, Review, Event. These have ongoing display value in Google Search and contribute meaningfully to AI understanding of your pages.
- Step 6: Tighten the underlying content. The FAQ deprecation rewards sites with substantive content over those leaning on display features. Audit your top-traffic pages for content depth, clarity, and direct-answer quality. Pages that already answer questions well don’t need the FAQ schema to keep performing.
This is also a useful moment to reassess how your business shows up across AI search surfaces. Our take on Google Search AI Mode tips for tracking and optimization covers the measurement side of that shift.
The Future of FAQ Structured Data
A few predictions worth holding loosely.
Google won’t bring it back. The two-step pattern, restrict eligibility, then retire the feature entirely, has now happened twice (HowTo, FAQ). Once Google takes a feature this far down the path, reversal is essentially unheard of.
Other rich result types are not at the same risk, yet. Product, Article, LocalBusiness, Review, and Event rich results have ongoing commercial value to Google and broader use across the SERP. The FAQ deprecation is not a signal that everything else is on the chopping block.
AI Overview citation patterns will keep evolving. The current best practice, clear prose, direct answers, structured headings, and entity clarity, will likely outlast any specific schema-type advice. If you optimize for human readability and crawler-parseable content structure, you stay aligned with where AI search is heading. We track the patterns we’re seeing in our breakdown of AI FAQs vs. FLUQs and how to win AI citations.
The schema type may quietly become more important for non-Google systems. Bing, Perplexity, ChatGPT search, and various agent crawlers continue to parse structured data. As discovery splits across more surfaces, sites with clean schema across more types will have a slight advantage. Slight, not transformative.
Practical Takeaways
- The FAQ rich result is gone. That’s the only operationally meaningful change.
- FAQPage schema is not deprecated. You can keep it, remove it, or refactor it without affecting Google rankings.
- The Search Console FAQ report disappears in June 2026. The API support follows in August. Export anything you need before those dates.
- Adding FAQ schema to a product or service page in 2026 will not lift your rankings. Stop using it as a generic SEO move.
Where your content is genuinely Q&A-shaped, support articles, regulator pages, help docs, guides built around predictable user questions, keep the markup and keep the content quality bar high.
The schema types still producing rich results (Product, Article, Organization, LocalBusiness, Review, Event) deserve more of your structured data attention now.
The bigger pattern matters: Google deprecates display features when they get gamed at scale. Building your SEO strategy around any single display feature is a fragile bet. Substance survives. Tactics don’t.
The teams that handle this update best won’t be the ones who remove FAQ schema fastest. They’ll be the ones who use the moment to step back, audit how much of their content strategy was leaning on display features, and rebuild around pages that genuinely deserve to rank, get cited, and earn clicks on their own merits.
Frequently Asked Questions
Only when your page genuinely contains a list of questions and answers users actually ask. After Google killed FAQ rich results on May 7, 2026, FAQ structured data no longer expands your SERP listing or drives a CTR lift. It also is not a documented signal for Google AI Overviews. Adding FAQPage markup to a product or service page for SEO reasons is no longer a meaningful tactic. Where the content is actually Q&A, support docs, help articles, regulator FAQs, keep the markup. Elsewhere, leave it off.
The Google Search Central FAQ structured data documentation still includes the technical specification for FAQPage markup, but it now opens with a deprecation banner stating that FAQ rich results no longer appear in Google Search as of May 7, 2026. The underlying schema definitions, required properties, and quality guidelines remain published. They describe what valid markup looks like, even though Google itself no longer surfaces a rich result for it.
Almost certainly not. Google has now followed the same two-step deprecation pattern with FAQ that it used for HowTo: restrict eligibility first, then retire the feature. Once Google moves a structured data feature this far through that process, it does not return. The strategically sound assumption is that FAQ rich results are gone permanently.
Most third-party FAQ rich results trackers stopped showing meaningful data after the 2023 eligibility restriction, and the few that continued tracking the government and health vertical will lose their data source when Google removes the FAQ search appearance filter from Search Console in June 2026. Internally, you can still pull FAQ-specific impression data from the Search Console API until August 2026, after that, the field is retired. Beyond that date, no reliable tracker exists.
Even with the rich result gone, the underlying Google FAQ rich results guidelines still describe what valid markup should look like: use FAQPage only when you provide single official answers (not user-submitted ones), include the full question and answer text in both the markup and the visible page, avoid promotional content inside Answer fields, and don’t use it to mark up content the user can’t actually see. Following these guidelines keeps your schema valid for other systems that still process it, including Bing, Perplexity, and various AI crawlers.
No. Google FAQ rich results eligibility in 2026 is effectively closed. Between August 2023 and May 2026, eligibility was technically restricted to well-known, authoritative government and health websites, but as of May 7, 2026, the feature itself stopped firing across all categories. There is no site type, however authoritative, that can earn a FAQ rich result in Google Search today.
No, not as a blanket move. Google has explicitly stated that unused structured data doesn’t cause Search problems, and FAQPage remains a valid schema.org type. Removing every FAQ schema block across a large site is engineering work for no measurable benefit. The smarter approach is selective: remove FAQ page structured data from pages where it was added decoratively to chase the rich result, and keep it on pages where the content is genuinely Q&A-shaped.
The evidence is not conclusive. Google has not connected the May 2026 deprecation to AI Overview inclusion, and commentary from senior practitioners, including Glenn Gabe, suggests Google’s actions imply FAQ markup is not weighted as a primary signal for AI surfaces. Pages do get cited in AI Overviews without any FAQ schema. The safer position is that clear prose, direct answers, and substantive content matter more for AI citation than the specific schema type.
FAQPage is for pages where the site owner provides a single official answer to each question, such as a help center entry or a regulator’s information page. QAPage is for pages where one question receives multiple user-submitted answers, such as a forum thread or community Q&A board. Choosing the wrong one means search engines and AI crawlers ignore the markup. The May 2026 deprecation only affects FAQPage rich results, QAPage rich results were always handled separately.
The deprecation affects how Google handles FAQ markup in its own surfaces, not how other AI search systems process it. Bing Copilot, ChatGPT search, Perplexity, and various AI crawlers continue to parse FAQPage markup as part of their general structured data handling. Whether they weigh it heavily or not is undocumented. The practical implication: if you’re optimizing for AI search, FAQ schema isn’t your highest-leverage move. Substantive, well-structured prose with clear questions as headings and complete answers underneath often achieves the same extraction outcome with less maintenance overhead.
